Added new model weights and info, trained on CV and CSS

.ipynb_checkpoints/ASR_Inference-checkpoint.ipynb

@@ -5,8 +5,8 @@
    "execution_count": 1,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:32:40.765119Z",
-     "start_time": "2021-03-14T19:32:39.314790Z"
+     "end_time": "2021-03-17T11:10:25.794375Z",
+     "start_time": "2021-03-17T11:10:24.301013Z"
     }
    },
    "outputs": [
@@ -36,8 +36,8 @@
    "execution_count": 2,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:32:40.774860Z",
-     "start_time": "2021-03-14T19:32:40.771235Z"
+     "end_time": "2021-03-17T11:10:29.608803Z",
+     "start_time": "2021-03-17T11:10:29.599700Z"
     }
    },
    "outputs": [],
@@ -75,11 +75,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:32:49.565850Z",
-     "start_time": "2021-03-14T19:32:41.891601Z"
+     "end_time": "2021-03-17T11:11:02.120225Z",
+     "start_time": "2021-03-17T11:10:56.182488Z"
     }
    },
    "outputs": [
@@ -98,11 +98,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 6,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:03.514113Z",
-     "start_time": "2021-03-14T19:33:00.953049Z"
+     "end_time": "2021-03-17T11:12:18.847005Z",
+     "start_time": "2021-03-17T11:12:14.919077Z"
     }
    },
    "outputs": [
@@ -110,8 +110,120 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Using custom data configuration el-afd0a157f05ee080\n",
-      "Reusing dataset common_voice (/home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564)\n"
+      "Using custom data configuration el-afd0a157f05ee080\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Downloading and preparing dataset common_voice/el (download: 363.89 MiB, generated: 4.75 MiB, post-processed: Unknown size, total: 368.64 MiB) to /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/0041e06ab061b91d0a23234a2221e87970a19cf3a81b20901474cffffeb7869f...\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r",
+      "Dataset common_voice downloaded and prepared to /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/0041e06ab061b91d0a23234a2221e87970a19cf3a81b20901474cffffeb7869f. Subsequent calls will reuse this data.\n"
      ]
     }
    ],
@@ -121,11 +233,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 7,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:03.528699Z",
-     "start_time": "2021-03-14T19:33:03.525034Z"
+     "end_time": "2021-03-17T11:12:18.860240Z",
+     "start_time": "2021-03-17T11:12:18.857252Z"
     }
    },
    "outputs": [],
@@ -135,19 +247,33 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 8,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:03.542260Z",
-     "start_time": "2021-03-14T19:33:03.538498Z"
+     "end_time": "2021-03-17T11:12:18.928497Z",
+     "start_time": "2021-03-17T11:12:18.869198Z"
     }
    },
    "outputs": [
     {
-     "name": "stderr",
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9869698af86e44bca75c4252996ff1a3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, max=1522), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-0ce2ebca66096fff.arrow\n"
+      "\n"
      ]
     }
    ],
@@ -157,19 +283,33 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 9,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:03.561798Z",
-     "start_time": "2021-03-14T19:33:03.554256Z"
+     "end_time": "2021-03-17T11:12:40.824595Z",
+     "start_time": "2021-03-17T11:12:18.937930Z"
     }
    },
    "outputs": [
     {
-     "name": "stderr",
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d232b2bb009543e0bb2542bce273c554",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, max=1522), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-38a09981767eff59.arrow\n"
+      "\n"
      ]
     }
    ],
@@ -179,11 +319,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 10,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:04.357229Z",
-     "start_time": "2021-03-14T19:33:03.570805Z"
+     "end_time": "2021-03-17T11:13:18.078738Z",
+     "start_time": "2021-03-17T11:12:40.834398Z"
     }
    },
    "outputs": [
@@ -191,112 +331,133 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-ba8c6dd59eb8ccf2.arrow\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-2e240883a5f827fd.arrow\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-485c00dc9048ed50.arrow\n"
+      "        "
      ]
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ffd787bc4ed048ae8f4977f2c539bedb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#0', max=191, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-44bf1791baae8e2e.arrow\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "79c51995d4f84ad8812230480d14b8cd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#2', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "52963d9cfd814346af070b2cc4e105cf",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#5', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-ecc0dfac5615a58e.arrow\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3b940160575143c7acfa142564e9f7d2",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#3', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "aa540f67ba894d7aa64e12fcdfab5ce0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#1', max=191, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-923d905502a8661d.arrow\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4962bdefdbbc44a7a44591480d8d6406",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#4', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e77f088bfe5644548fe2c4277d0c86da",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#7', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-bb54bb00dae79669.arrow\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5827f93e99994fe9919aac53f0fb9444",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#6', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-062aeafc3b8816c1.arrow\n"
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n"
      ]
     }
    ],
@@ -306,11 +467,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 11,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:11.205598Z",
-     "start_time": "2021-03-14T19:33:04.368615Z"
+     "end_time": "2021-03-17T11:13:25.145155Z",
+     "start_time": "2021-03-17T11:13:18.091929Z"
     }
    },
    "outputs": [
@@ -332,12 +493,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "b179696d93284b739cc550511ca28b78",
+       "model_id": "ae326a173a044b1494793e2a70d76a87",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#1', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#0', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -346,12 +507,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "2aa538a2fdcb4d56a55cd612c879044f",
+       "model_id": "21ab1ef2af5a4a4fb23c68b0c5cf32f8",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#0', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#1', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -360,12 +521,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "073b156d7b7541e08ad20ec377efb05b",
+       "model_id": "d331c5f4f888477daceffe370f6cd89f",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#2', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#3', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -374,12 +535,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "dcaa10621a674699a7bf17357cd75a4a",
+       "model_id": "6fa790118aa340e4afb9f83e71403a13",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#4', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#2', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -388,12 +549,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "9ee2184681db4e16802387bbc1acfef3",
+       "model_id": "c8092e2f59a9404596dc2bab206edf2c",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#3', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#5', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -402,12 +563,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "36c3bcde96114167a1dc6942d9d551f1",
+       "model_id": "20f913f0caf8401098743b9e5051fc52",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#6', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#4', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -416,12 +577,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "b0ad190f479d40a599e74cfa9deddec8",
+       "model_id": "7c7e15e24384494cb49a72106ce41ccd",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#5', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#6', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -430,7 +591,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "138f7261158c4a9ab84414a20954de01",
+       "model_id": "73245add55e24ee2a6dbe0713d5073d9",
        "version_major": 2,
        "version_minor": 0
       },
@@ -462,11 +623,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 12,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:14.391497Z",
-     "start_time": "2021-03-14T19:33:11.216118Z"
+     "end_time": "2021-03-17T11:14:12.721500Z",
+     "start_time": "2021-03-17T11:14:08.198478Z"
     }
    },
    "outputs": [
@@ -474,8 +635,120 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Using custom data configuration el-ac779bf2c9f7c09b\n",
-      "Reusing dataset common_voice (/home/earendil/.cache/huggingface/datasets/common_voice/el-ac779bf2c9f7c09b/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564)\n"
+      "Using custom data configuration el-ac779bf2c9f7c09b\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Downloading and preparing dataset common_voice/el (download: 363.89 MiB, generated: 4.75 MiB, post-processed: Unknown size, total: 368.64 MiB) to /home/earendil/.cache/huggingface/datasets/common_voice/el-ac779bf2c9f7c09b/6.1.0/0041e06ab061b91d0a23234a2221e87970a19cf3a81b20901474cffffeb7869f...\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r",
+      "Dataset common_voice downloaded and prepared to /home/earendil/.cache/huggingface/datasets/common_voice/el-ac779bf2c9f7c09b/6.1.0/0041e06ab061b91d0a23234a2221e87970a19cf3a81b20901474cffffeb7869f. Subsequent calls will reuse this data.\n"
      ]
     }
    ],
@@ -536,6 +809,96 @@
     "# πού θέλεις να πάμε; ρώτησε φοβισμένα ο βασιλιάς."
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2021-03-17T11:15:35.637739Z",
+     "start_time": "2021-03-17T11:14:14.689842Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1f7ba9e12187401f870555d20a6a9458",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, max=1522), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "def map_to_result(batch):\n",
+    "  model.to(\"cuda\")\n",
+    "  input_values = processor(\n",
+    "      batch[\"input_values\"], \n",
+    "      sampling_rate=16_000, \n",
+    "      return_tensors=\"pt\"\n",
+    "  ).input_values.to(\"cuda\")\n",
+    "\n",
+    "  with torch.no_grad():\n",
+    "    logits = model(input_values).logits\n",
+    "\n",
+    "  pred_ids = torch.argmax(logits, dim=-1)\n",
+    "  batch[\"pred_str\"] = processor.batch_decode(pred_ids)[0]\n",
+    "\n",
+    "  return batch\n",
+    "\n",
+    "results = common_voice_test.map(map_to_result)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2021-03-17T11:17:11.951524Z",
+     "start_time": "2021-03-17T11:17:08.856552Z"
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test WER: 0.396\n"
+     ]
+    }
+   ],
+   "source": [
+    "def compute_metrics(pred):\n",
+    "    pred_logits = pred.predictions\n",
+    "    pred_ids = np.argmax(pred_logits, axis=-1)\n",
+    "\n",
+    "    pred.label_ids[pred.label_ids == -100] = processor.tokenizer.pad_token_id\n",
+    "\n",
+    "    pred_str = processor.batch_decode(pred_ids)\n",
+    "    # we do not want to group tokens when computing the metrics\n",
+    "    label_str = processor.batch_decode(pred.label_ids, group_tokens=False)\n",
+    "\n",
+    "    wer = wer_metric.compute(predictions=pred_str, references=label_str)\n",
+    "\n",
+    "    return {\"wer\": wer}\n",
+    "\n",
+    "wer_metric = load_metric(\"wer\")\n",
+    "\n",
+    "print(\"Test WER: {:.3f}\".format(wer_metric.compute(predictions=results[\"pred_str\"], references= [item.lower() for item in common_voice_test_transcription['sentence']])))"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,

ASR_Inference.ipynb

@@ -5,8 +5,8 @@
    "execution_count": 1,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:32:40.765119Z",
-     "start_time": "2021-03-14T19:32:39.314790Z"
+     "end_time": "2021-03-17T11:10:25.794375Z",
+     "start_time": "2021-03-17T11:10:24.301013Z"
     }
    },
    "outputs": [
@@ -36,8 +36,8 @@
    "execution_count": 2,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:32:40.774860Z",
-     "start_time": "2021-03-14T19:32:40.771235Z"
+     "end_time": "2021-03-17T11:10:29.608803Z",
+     "start_time": "2021-03-17T11:10:29.599700Z"
     }
    },
    "outputs": [],
@@ -75,11 +75,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:32:49.565850Z",
-     "start_time": "2021-03-14T19:32:41.891601Z"
+     "end_time": "2021-03-17T11:11:02.120225Z",
+     "start_time": "2021-03-17T11:10:56.182488Z"
     }
    },
    "outputs": [
@@ -98,11 +98,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 6,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:03.514113Z",
-     "start_time": "2021-03-14T19:33:00.953049Z"
+     "end_time": "2021-03-17T11:12:18.847005Z",
+     "start_time": "2021-03-17T11:12:14.919077Z"
     }
    },
    "outputs": [
@@ -110,8 +110,120 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Using custom data configuration el-afd0a157f05ee080\n",
-      "Reusing dataset common_voice (/home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564)\n"
+      "Using custom data configuration el-afd0a157f05ee080\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Downloading and preparing dataset common_voice/el (download: 363.89 MiB, generated: 4.75 MiB, post-processed: Unknown size, total: 368.64 MiB) to /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/0041e06ab061b91d0a23234a2221e87970a19cf3a81b20901474cffffeb7869f...\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r",
+      "Dataset common_voice downloaded and prepared to /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/0041e06ab061b91d0a23234a2221e87970a19cf3a81b20901474cffffeb7869f. Subsequent calls will reuse this data.\n"
      ]
     }
    ],
@@ -121,11 +233,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 7,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:03.528699Z",
-     "start_time": "2021-03-14T19:33:03.525034Z"
+     "end_time": "2021-03-17T11:12:18.860240Z",
+     "start_time": "2021-03-17T11:12:18.857252Z"
     }
    },
    "outputs": [],
@@ -135,19 +247,33 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 8,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:03.542260Z",
-     "start_time": "2021-03-14T19:33:03.538498Z"
+     "end_time": "2021-03-17T11:12:18.928497Z",
+     "start_time": "2021-03-17T11:12:18.869198Z"
     }
    },
    "outputs": [
     {
-     "name": "stderr",
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9869698af86e44bca75c4252996ff1a3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, max=1522), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-0ce2ebca66096fff.arrow\n"
+      "\n"
      ]
     }
    ],
@@ -157,19 +283,33 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 9,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:03.561798Z",
-     "start_time": "2021-03-14T19:33:03.554256Z"
+     "end_time": "2021-03-17T11:12:40.824595Z",
+     "start_time": "2021-03-17T11:12:18.937930Z"
     }
    },
    "outputs": [
     {
-     "name": "stderr",
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d232b2bb009543e0bb2542bce273c554",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, max=1522), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-38a09981767eff59.arrow\n"
+      "\n"
      ]
     }
    ],
@@ -179,11 +319,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 10,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:04.357229Z",
-     "start_time": "2021-03-14T19:33:03.570805Z"
+     "end_time": "2021-03-17T11:13:18.078738Z",
+     "start_time": "2021-03-17T11:12:40.834398Z"
     }
    },
    "outputs": [
@@ -191,112 +331,133 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-ba8c6dd59eb8ccf2.arrow\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-2e240883a5f827fd.arrow\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-485c00dc9048ed50.arrow\n"
+      "        "
      ]
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ffd787bc4ed048ae8f4977f2c539bedb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#0', max=191, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-44bf1791baae8e2e.arrow\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "79c51995d4f84ad8812230480d14b8cd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#2', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "52963d9cfd814346af070b2cc4e105cf",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#5', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-ecc0dfac5615a58e.arrow\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3b940160575143c7acfa142564e9f7d2",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#3', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "aa540f67ba894d7aa64e12fcdfab5ce0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#1', max=191, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-923d905502a8661d.arrow\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4962bdefdbbc44a7a44591480d8d6406",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#4', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e77f088bfe5644548fe2c4277d0c86da",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#7', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-bb54bb00dae79669.arrow\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5827f93e99994fe9919aac53f0fb9444",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, description='#6', max=190, style=ProgressStyle(description_width='initial'…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-062aeafc3b8816c1.arrow\n"
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n"
      ]
     }
    ],
@@ -306,11 +467,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 11,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:11.205598Z",
-     "start_time": "2021-03-14T19:33:04.368615Z"
+     "end_time": "2021-03-17T11:13:25.145155Z",
+     "start_time": "2021-03-17T11:13:18.091929Z"
     }
    },
    "outputs": [
@@ -332,12 +493,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "b179696d93284b739cc550511ca28b78",
+       "model_id": "ae326a173a044b1494793e2a70d76a87",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#1', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#0', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -346,12 +507,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "2aa538a2fdcb4d56a55cd612c879044f",
+       "model_id": "21ab1ef2af5a4a4fb23c68b0c5cf32f8",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#0', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#1', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -360,12 +521,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "073b156d7b7541e08ad20ec377efb05b",
+       "model_id": "d331c5f4f888477daceffe370f6cd89f",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#2', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#3', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -374,12 +535,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "dcaa10621a674699a7bf17357cd75a4a",
+       "model_id": "6fa790118aa340e4afb9f83e71403a13",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#4', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#2', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -388,12 +549,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "9ee2184681db4e16802387bbc1acfef3",
+       "model_id": "c8092e2f59a9404596dc2bab206edf2c",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#3', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#5', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -402,12 +563,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "36c3bcde96114167a1dc6942d9d551f1",
+       "model_id": "20f913f0caf8401098743b9e5051fc52",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#6', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#4', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -416,12 +577,12 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "b0ad190f479d40a599e74cfa9deddec8",
+       "model_id": "7c7e15e24384494cb49a72106ce41ccd",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "HBox(children=(IntProgress(value=0, description='#5', max=24, style=ProgressStyle(description_width='initial')…"
+       "HBox(children=(IntProgress(value=0, description='#6', max=24, style=ProgressStyle(description_width='initial')…"
       ]
      },
      "metadata": {},
@@ -430,7 +591,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "138f7261158c4a9ab84414a20954de01",
+       "model_id": "73245add55e24ee2a6dbe0713d5073d9",
        "version_major": 2,
        "version_minor": 0
       },
@@ -462,11 +623,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 12,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T19:33:14.391497Z",
-     "start_time": "2021-03-14T19:33:11.216118Z"
+     "end_time": "2021-03-17T11:14:12.721500Z",
+     "start_time": "2021-03-17T11:14:08.198478Z"
     }
    },
    "outputs": [
@@ -474,8 +635,120 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Using custom data configuration el-ac779bf2c9f7c09b\n",
-      "Reusing dataset common_voice (/home/earendil/.cache/huggingface/datasets/common_voice/el-ac779bf2c9f7c09b/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564)\n"
+      "Using custom data configuration el-ac779bf2c9f7c09b\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Downloading and preparing dataset common_voice/el (download: 363.89 MiB, generated: 4.75 MiB, post-processed: Unknown size, total: 368.64 MiB) to /home/earendil/.cache/huggingface/datasets/common_voice/el-ac779bf2c9f7c09b/6.1.0/0041e06ab061b91d0a23234a2221e87970a19cf3a81b20901474cffffeb7869f...\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=1, bar_style='info', max=1), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\r",
+      "Dataset common_voice downloaded and prepared to /home/earendil/.cache/huggingface/datasets/common_voice/el-ac779bf2c9f7c09b/6.1.0/0041e06ab061b91d0a23234a2221e87970a19cf3a81b20901474cffffeb7869f. Subsequent calls will reuse this data.\n"
      ]
     }
    ],
@@ -536,6 +809,96 @@
     "# πού θέλεις να πάμε; ρώτησε φοβισμένα ο βασιλιάς."
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2021-03-17T11:15:35.637739Z",
+     "start_time": "2021-03-17T11:14:14.689842Z"
+    }
+   },
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1f7ba9e12187401f870555d20a6a9458",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, max=1522), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "def map_to_result(batch):\n",
+    "  model.to(\"cuda\")\n",
+    "  input_values = processor(\n",
+    "      batch[\"input_values\"], \n",
+    "      sampling_rate=16_000, \n",
+    "      return_tensors=\"pt\"\n",
+    "  ).input_values.to(\"cuda\")\n",
+    "\n",
+    "  with torch.no_grad():\n",
+    "    logits = model(input_values).logits\n",
+    "\n",
+    "  pred_ids = torch.argmax(logits, dim=-1)\n",
+    "  batch[\"pred_str\"] = processor.batch_decode(pred_ids)[0]\n",
+    "\n",
+    "  return batch\n",
+    "\n",
+    "results = common_voice_test.map(map_to_result)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2021-03-17T11:17:11.951524Z",
+     "start_time": "2021-03-17T11:17:08.856552Z"
+    }
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test WER: 0.396\n"
+     ]
+    }
+   ],
+   "source": [
+    "def compute_metrics(pred):\n",
+    "    pred_logits = pred.predictions\n",
+    "    pred_ids = np.argmax(pred_logits, axis=-1)\n",
+    "\n",
+    "    pred.label_ids[pred.label_ids == -100] = processor.tokenizer.pad_token_id\n",
+    "\n",
+    "    pred_str = processor.batch_decode(pred_ids)\n",
+    "    # we do not want to group tokens when computing the metrics\n",
+    "    label_str = processor.batch_decode(pred.label_ids, group_tokens=False)\n",
+    "\n",
+    "    wer = wer_metric.compute(predictions=pred_str, references=label_str)\n",
+    "\n",
+    "    return {\"wer\": wer}\n",
+    "\n",
+    "wer_metric = load_metric(\"wer\")\n",
+    "\n",
+    "print(\"Test WER: {:.3f}\".format(wer_metric.compute(predictions=results[\"pred_str\"], references= [item.lower() for item in common_voice_test_transcription['sentence']])))"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,

README.md

@@ -21,7 +21,7 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 31.05
+         value: 10.497628
 ---
 
 # Greek (el) version of the XLSR-Wav2Vec2 automatic speech recognition (ASR) model
@@ -29,12 +29,14 @@ model-index:
 
 * language: el
 * licence: apache-2.0
-* dataset: CommonVoice (EL), 364MB: https://commonvoice.mozilla.org/el/datasets 
-* model: XLSR-Wav2Vec2, trained for 60 epochs
+* dataset: CommonVoice (EL), 364MB: https://commonvoice.mozilla.org/el/datasets + CSS10 (EL), 1.22GB: https://github.com/Kyubyong/css10
+* model: XLSR-Wav2Vec2, trained for 50 epochs
 * metrics: Word Error Rate (WER)
 
 ## Model description
 
+UPDATE: We repeated the fine-tuning process using an additional 1.22GB dataset from CSS10.
+
 Wav2Vec2 is a pretrained model for Automatic Speech Recognition (ASR) and was released in September 2020 by Alexei Baevski, Michael Auli, and Alex Conneau. Soon after the superior performance of Wav2Vec2 was demonstrated on the English ASR dataset LibriSpeech, Facebook AI presented XLSR-Wav2Vec2. XLSR stands for cross-lingual speech representations and refers to XLSR-Wav2Vec2`s ability to learn speech representations that are useful across multiple languages.
 
 Similar to Wav2Vec2, XLSR-Wav2Vec2 learns powerful speech representations from hundreds of thousands of hours of speech in more than 50 languages of unlabeled speech. Similar, to BERT's masked language modeling, the model learns contextualized speech representations by randomly masking feature vectors before passing them to a transformer network.
@@ -189,7 +191,7 @@ result = test_dataset.map(evaluate, batched=True, batch_size=8)
 print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
 ```
 
-**Test Result**: 31.05 % 
+**Test Result**: 10.497628 % 
 
 ### How to use for training:
 
@@ -200,9 +202,9 @@ Instructions and code to replicate the process are provided in the Fine_Tune_XLS
 
 | Metric      | Value |
 | ----------- | ----------- |
-| Training Loss | 0.0287 |
+| Training Loss | 0.0545 |
 | Validation Loss | 0.6062 |
-| WER on CommonVoice Test *| 0.3105 |
+| WER on CommonVoice Test (%) *| 10.497628 |
 * Reference transcripts were lower-cased and striped of punctuation and special characters.
 
 Full metrics log here: